A portion of the disclosure of this patent document contains material, which is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure, as it appears in the Patent and Trademark Office patent file or records, but otherwise reserves all copyright rights whatsoever.
The present invention relates to optical image processing, and in particular to flashlamp circuits for stroboscopic illumination of objects.
Modern digital computing technology is being called upon to perform an ever-increasing variety of tasks. Machines, which once responded purely to manual direction, are now being equipped with computer processors, enabling them to assist a human operator. Manufacturing lines, which produce volumes of standardized assemblies, are being equipped with computer-controlled process machinery. Industrial robots have the capability of being re-programmed to perform many different tasks within the mechanical limits of motion of the device.
The automated analysis of captured optical images has great utility for digital control systems. For example, optical imaging may be used to great advantage in automated manufacturing environments, although this is not necessarily the only possible application. Usually, the acquisition of optical images does not interfere with sensitive parts or manufacturing processes, as other forms of measurement might. Optical images of manufactured articles may be captured and analyzed for purposes of inspection, or for guiding the motion of process machinery, such as an industrial robot, relative to a workpiece.
In many applications, it is desirable to create a height image or profile of a target object, in order to produce a 2-dimensional map of surface heights. One particular example of this is the inspection of solder deposits on electronic printed circuit cards. As well appreciated by those knowledgeable in the industry, at an intermediate stage of manufacture, these cards may have hundreds or thousands of small solder deposits, which are electrically coupled to circuit paths printed within the card. When electrical components are later mounted on the card, the solder is melted to form electrical connections between the circuit paths in the card and pins, wires, or other conductors from the components. The increasing complexity of the information age demands that these components have larger and larger numbers of connections, usually within smaller and smaller areas. An insufficient amount of solder at a connection site may result in a failure to make the connection, or a connection that intermittently fails or fails after some time in the field. Excess solder or misplaced solder can similarly wreak havoc with the resulting product. The size and number of such connections places great demands on the consistency of the manufacturing process. It also makes it difficult to inspect a card for defects. At the same time, the cost of an undetected defect can be large. Accordingly, there is substantial potential benefit in an automated process, which can accurately inspect solder deposits quickly and without damage to the card. A height profile of a circuit card with solder deposits, taken from optical measurements, can be used to determine the volume of solder at each connection site.
One technique for generating a height profile of a target object from optical measurements is known as phase profilometry. In this technique, light illuminates the target object and at least two images of the target object are acquired, each image acquired either at different phases of light, or at differing positions of the target. In either event, a phase shift is introduced between any two of the images. The images are then combined by image processing techniques to reconstruct a height image. Various methods for phase profilometry are disclosed in U.S. Pat. Nos. 4,657,394, 4,641,972, 5,636,025, 5,646,733 and 6,049,384.
The technical problem of capturing at least two images of a target object is non-trivial. It is desirable to capture the images in rapid succession, in order to reduce mis-registration caused by undesired motion between the different exposures, and support a high throughput of image capture and analysis. In particular, it is desirable to wait no more than 1 millisecond between any two successive image acquisitions to be combined. While it may be possible to generate successive images within approximately 1 millisecond or less using existing techniques, such techniques involve excessive power consumption and/or excessive hardware, or involve other undesirable side effects. For example, in the case of three-phase profilometry, it is possible to replicate three separate lamps, circuits, and associated hardware for acquiring three separate images, but this would involve considerable hardware expense, and would introduce additional variables if the illumination from different sources were not identical. Additionally, the peak power consumption for known circuits that discharge a single flashlamp with approximately 1 millisecond spacing is typically on the order of 200 watts, which is beyond the capabilities of known small high-voltage (HV) supplies.
Techniques have been proposed that reduce power consumption and/or excessive hardware by providing a resonant charging circuit that charges a discharge capacitor from a large reservoir capacitor. An example of such teaching is set forth in U.S. Pat. No. 3,953,763 to Herrick. The inherent dynamics of the circuit of Herrick allow the discharge capacitor to be charged to roughly twice the voltage of the reservoir capacitor. Such resonant charging is accomplished with low dissipation. While the circuit of Herrick provides a number of advantages, it is not without need for improvement. For example, aspects of the Herrick circuit are believed to have unduly shortened the lifetime of a tested flashlamp. The circuit of Herrick cannot be used without an inductor, because without adequate inductance in the circuit, the di/dt of the circuit would exceed the maximum allowable for most commercially available SCRs, causing SCR failure from internal hotspots. Here i denotes current and t denotes time. Addition of an inductor can relieve this problem, since the di/dt is limited to approximately v/L, where v is the discharge potential and L is the inductance. For typical SCRs, the di/dt limit of 200 A/xcexcs, together with the 450-V discharge potential, indicates that an inductor of at least 2 xcexcH is needed. This value of inductance significantly lengthens the tail of the discharge, which has the disadvantage of shortening lamp life. Although SCR devices are available with higher di/dt ratings than the usual 200 A/xcexcs, they are expensive and prohibitively bulky.
Further, the circuit does not provide for a fast, convenient discharge of the reservoir capacitor for safety in handling and repairing the circuit. Finally, the circuit of Herrick does not provide selectable discharge energies. A rapid firing flashlamp discharge circuit providing resonant charging and addressing the limitations above thus provides a significant improvement.
A flashlamp circuit includes a charge reservoir that receives a first voltage from an external source. The charge reservoir is coupled to a resonator and a plurality of discharge capacitors to provide a second voltage to the plurality of discharge capacitors that is greater than the first voltage. A switch is disposed between at least one of the discharge capacitors and ground to selectively charge the at least one discharge capacitor based upon an input to the switch. Discharge energy is passed from the discharge capacitor(s) to a flashlamp through a discharge bank without passing through any inductive elements. A bleeder circuit can be interposed between the power supply and the reservoir to discharge the reservoir upon shutdown.